CN110428011A - A kind of deep learning image fault classification method towards video transmission quality - Google Patents
A kind of deep learning image fault classification method towards video transmission quality Download PDFInfo
- Publication number
- CN110428011A CN110428011A CN201910721496.4A CN201910721496A CN110428011A CN 110428011 A CN110428011 A CN 110428011A CN 201910721496 A CN201910721496 A CN 201910721496A CN 110428011 A CN110428011 A CN 110428011A
- Authority
- CN
- China
- Prior art keywords
- image
- training
- function
- deep learning
- transmission quality
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000005540 biological transmission Effects 0.000 title claims abstract description 16
- 238000013135 deep learning Methods 0.000 title claims abstract description 15
- 238000012549 training Methods 0.000 claims abstract description 37
- 230000006870 function Effects 0.000 claims abstract description 36
- 230000002708 enhancing effect Effects 0.000 claims abstract description 8
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 6
- 230000008569 process Effects 0.000 claims abstract description 6
- 239000012141 concentrate Substances 0.000 claims abstract description 4
- 238000013461 design Methods 0.000 claims abstract description 4
- 230000001537 neural effect Effects 0.000 claims abstract description 4
- 238000012545 processing Methods 0.000 claims abstract description 4
- 230000006835 compression Effects 0.000 claims description 12
- 238000007906 compression Methods 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 2
- 230000008929 regeneration Effects 0.000 claims 1
- 238000011069 regeneration method Methods 0.000 claims 1
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The deep learning image fault classification method towards video transmission quality that the invention discloses a kind of selects training loss function, makes training dataset this method comprises: building convolutional neural networks, are arranged training hyper parameter;It concentrates undistorted image to do data enhancing processing data, forms the training sample of more different type of distortion;The method of training sample is randomly selected in design batch training process;Complete neural metwork training, obtain training pattern, after actual deployment can real-time grading video flowing input target image type of distortion.
Description
Technical field
The present invention relates to image fault evaluation of classification field more particularly to a kind of image fault classification based on deep learning
Method.
Background technique
Video image can generate distortion because of various reasons in transmission process, so for occurring in image transmitting process
Distortion correctly classify just and seem critically important.Existing image fault sorting technique much all be rely on human visual system or from
The computation model of right image statistics, even manual sort.It is relatively low that the former is distorted classification accuracy, is easy to appear judgement not
Unanimous circumstances;And then efficiency is very low by the latter, working long hours is easy to appear fatigue error, influences judgment accuracy, seeks thus
A kind of method that can efficiently, accurately, intelligently classify transmission of video images distortion is looked for have important practical significance.
Summary of the invention
In order to solve the above technical problems, the object of the present invention is to provide a kind of deep learning figure towards video transmission quality
Image distortion classification method.
The purpose of the present invention is realized by technical solution below:
A kind of deep learning image fault classification method towards video transmission quality, comprising:
A constructs convolutional neural networks, and training hyper parameter is arranged, and selects training loss function Loss, makes training dataset;
B concentrates undistorted image to do data enhancing processing data, forms the training sample of more different type of distortion;
The method of training sample is randomly selected in C design batch training process;
D completes neural metwork training, obtains training pattern, after actual deployment can the input of real-time grading video flowing target
Image fault type.
Detailed description of the invention
Fig. 1 is the deep learning image fault classification method flow chart towards video transmission quality.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with examples and drawings to this hair
It is bright to be described in further detail.
As shown in Figure 1, a kind of deep learning image fault classification method process towards video transmission quality, including it is following
Step:
Step 10 constructs convolutional neural networks, and training hyper parameter is arranged, and selects training loss function Loss, makes training number
According to collection;
Step 20 concentrates undistorted image to do data enhancing processing data, forms the training sample of more different type of distortion
This;
The method of training sample is randomly selected in step 30 design batch training process;
Step 40 completes neural metwork training, obtains training pattern, can the input of real-time grading video flowing after actual deployment
Target image type of distortion;
Above-mentioned steps 10 specifically include: hyper parameter setting includes maximum train epochs S, single batch number of training N, initial
Learning rate η0, learning rate attenuation rate d, learning rate update interval M, input image size etc..
Above-mentioned steps 10 specifically include: use softmax function as the classifier of CNN, select intersect entropy function as
Loss function.
Above-mentioned steps 10 specifically include: note ηtThe learning rate at interval is updated for the t times study,For the t times study update
Interim least disadvantage functional value, LminFor global minima loss function value, then ηtUpdate method are as follows:
Above-mentioned steps 10 specifically include: altogether including undistorted C in data setc, white noise sound distortion Cn, fuzzy distortion Cb、JPEG
Compression artefacts CJAnd JPEG2000 compression artefacts CJ2000Totally 5 seed type image, and wherein 60% sample is randomly selected from data set
This is as training set, 20% sample as test set, 20% sample as verifying collection.
Above-mentioned steps 20 specifically include: note I is undistorted image, θnWhite noise function parameter, f are added for imagen(I,θn)
White noise function, θ are added for imagebAmbiguity function parameter, f are added for imageb(I,θb) it is that image adds ambiguity function, θJFor
Image JPEG compression function parameter, fJ(I,θJ) it is image JPEG compression function, θJ2000For image JPEG2000 compression function ginseng
Number, fJ2000(I,θJ2000) it is image JPEG2000 compression function, by that can realize in addition to undistorted type other to above-mentioned function
The data of type enhance.
Above-mentioned steps 30 specifically include: note I*Image, C to randomly select are to randomly select image type, fC() is
Data enhancing function, the θ of corresponding typesCFor function fCRandom parameter, random (C) within the scope of () reasonable value are data increasing
Strong random selection function, then I*Are as follows:
Although disclosed herein embodiment it is as above, the content is only to facilitate understanding the present invention and adopting
Embodiment is not intended to limit the invention.Any those skilled in the art to which this invention pertains are not departing from this
Under the premise of the disclosed spirit and scope of invention, any modification and change can be made in the implementing form and in details,
But scope of patent protection of the invention, still should be subject to the scope of the claims as defined in the appended claims.
Claims (7)
1. a kind of deep learning image fault classification method towards video transmission quality, which is characterized in that the described method includes:
A constructs convolutional neural networks, and training hyper parameter is arranged, and selects training loss function Loss, makes training dataset;
B concentrates undistorted image to do data enhancing processing data, forms the training sample of more different type of distortion;
The method of training sample is randomly selected in C design batch training process;
D completes neural metwork training, obtains training pattern, after actual deployment can the input of real-time grading video flowing target image
Type of distortion.
2. the deep learning image fault classification method towards video transmission quality as described in claim 1, which is characterized in that
In the step A, the setting of training hyper parameter includes maximum train epochs S, single batch number of training N, initial learning rate η0, learn
Habit rate attenuation rate d, learning rate update interval M and input image size.
3. the deep learning image fault classification method towards video transmission quality as described in claim 1, which is characterized in that
The step A is specifically included, and uses softmax function as the classifier of CNN, selects to intersect entropy function as loss function.
4. the deep learning image fault classification method towards video transmission quality as described in claim 1, which is characterized in that
In the step A, η is rememberedtThe learning rate at interval is updated for the t times study,The minimum damage between the t times study regeneration interval
Lose functional value, LminFor global minima loss function value, then ηtUpdate method are as follows:
5. the deep learning image fault classification method towards video transmission quality as described in claim 1, which is characterized in that
It altogether include undistorted C in data set in the step Ac, white noise sound distortion Cn, fuzzy distortion Cb, JPEG compression be distorted CJAnd
JPEG2000 compression artefacts CJ2000Totally 5 seed type image, and 60% sample composing training collection, 20% sample are extracted from data set
Constitute test set, 20% sample constitutes verifying collection.
6. the deep learning image fault classification method towards video transmission quality as described in claim 1, which is characterized in that
The step B is specifically included, and note I is undistorted image, θnWhite noise function parameter, f are added for imagen(I,θn) add for image
Add white noise function, θbAmbiguity function parameter, f are added for imageb(I,θb) it is that image adds ambiguity function, θJFor image JPEG
Compression function parameter, fJ(I,θJ) it is image JPEG compression function, θJ2000For image JPEG2000 compression function parameter, fJ2000
(I,θJ2000) it is image JPEG2000 compression function, by realizing the other kinds of data in addition to undistorted type to above-mentioned function
Enhancing.
7. the deep learning image fault classification method towards video transmission quality as described in claim 1, which is characterized in that
The step C is specifically included, and remembers I*Image, C to randomly select are to randomly select image type, fC() is corresponding types
Data enhancing function, θCFor function fCRandom parameter, random (C) within the scope of () reasonable value are data enhancing random selection
Function, then I*Are as follows:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910721496.4A CN110428011A (en) | 2019-08-06 | 2019-08-06 | A kind of deep learning image fault classification method towards video transmission quality |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910721496.4A CN110428011A (en) | 2019-08-06 | 2019-08-06 | A kind of deep learning image fault classification method towards video transmission quality |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110428011A true CN110428011A (en) | 2019-11-08 |
Family
ID=68412788
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910721496.4A Pending CN110428011A (en) | 2019-08-06 | 2019-08-06 | A kind of deep learning image fault classification method towards video transmission quality |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110428011A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111815529A (en) * | 2020-06-30 | 2020-10-23 | 上海电力大学 | Low-quality image classification enhancement method based on model fusion and data enhancement |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110274361A1 (en) * | 2010-05-10 | 2011-11-10 | Board Of Regents, The University Of Texas System | Determining quality of an image or video using a distortion classifier |
CN105208374A (en) * | 2015-08-24 | 2015-12-30 | 宁波大学 | Non-reference image quality objective evaluation method based on deep learning |
CN107770517A (en) * | 2017-10-24 | 2018-03-06 | 天津大学 | Full reference image quality appraisement method based on image fault type |
CN109215028A (en) * | 2018-11-06 | 2019-01-15 | 福州大学 | A kind of multiple-objection optimization image quality measure method based on convolutional neural networks |
CN109741315A (en) * | 2018-12-29 | 2019-05-10 | 中国传媒大学 | A kind of non-reference picture assessment method for encoding quality based on deeply study |
CN109816625A (en) * | 2018-11-27 | 2019-05-28 | 广东电网有限责任公司 | A kind of video quality score implementation method |
WO2019112084A1 (en) * | 2017-12-06 | 2019-06-13 | 한국과학기술원 | Method for removing compression distortion by using cnn |
CN109919025A (en) * | 2019-01-30 | 2019-06-21 | 华南理工大学 | Video scene Method for text detection, system, equipment and medium based on deep learning |
CN110084757A (en) * | 2019-04-15 | 2019-08-02 | 南京信息工程大学 | A kind of infrared depth image enhancement method based on generation confrontation network |
-
2019
- 2019-08-06 CN CN201910721496.4A patent/CN110428011A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110274361A1 (en) * | 2010-05-10 | 2011-11-10 | Board Of Regents, The University Of Texas System | Determining quality of an image or video using a distortion classifier |
CN105208374A (en) * | 2015-08-24 | 2015-12-30 | 宁波大学 | Non-reference image quality objective evaluation method based on deep learning |
CN107770517A (en) * | 2017-10-24 | 2018-03-06 | 天津大学 | Full reference image quality appraisement method based on image fault type |
WO2019112084A1 (en) * | 2017-12-06 | 2019-06-13 | 한국과학기술원 | Method for removing compression distortion by using cnn |
CN109215028A (en) * | 2018-11-06 | 2019-01-15 | 福州大学 | A kind of multiple-objection optimization image quality measure method based on convolutional neural networks |
CN109816625A (en) * | 2018-11-27 | 2019-05-28 | 广东电网有限责任公司 | A kind of video quality score implementation method |
CN109741315A (en) * | 2018-12-29 | 2019-05-10 | 中国传媒大学 | A kind of non-reference picture assessment method for encoding quality based on deeply study |
CN109919025A (en) * | 2019-01-30 | 2019-06-21 | 华南理工大学 | Video scene Method for text detection, system, equipment and medium based on deep learning |
CN110084757A (en) * | 2019-04-15 | 2019-08-02 | 南京信息工程大学 | A kind of infrared depth image enhancement method based on generation confrontation network |
Non-Patent Citations (2)
Title |
---|
刘桂雄等: ""运动模糊图像复原技术研究进展与展望"", 《激光杂志》 * |
刘桂雄等: ""运动模糊图像复原技术研究进展与展望"", 《激光杂志》, vol. 40, no. 4, 25 April 2019 (2019-04-25), pages 1 - 6 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111815529A (en) * | 2020-06-30 | 2020-10-23 | 上海电力大学 | Low-quality image classification enhancement method based on model fusion and data enhancement |
CN111815529B (en) * | 2020-06-30 | 2023-02-07 | 上海电力大学 | Low-quality image classification enhancement method based on model fusion and data enhancement |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108520504A (en) | A kind of blurred picture blind restoration method based on generation confrontation network end-to-end | |
US11176418B2 (en) | Model test methods and apparatuses | |
CN106682704A (en) | Method of disease image identification based on hybrid convolutional neural network fused with context information | |
CN106780466A (en) | A kind of cervical cell image-recognizing method based on convolutional neural networks | |
CN109118445B (en) | Underwater image enhancement method based on multi-branch generation countermeasure network | |
CN108665460A (en) | Image quality evaluating method based on combination neural net and Classification Neural | |
CN107239514A (en) | A kind of plants identification method and system based on convolutional neural networks | |
CN107689034A (en) | A kind of training method of neutral net, denoising method and device | |
CN110135386B (en) | Human body action recognition method and system based on deep learning | |
CN114092769B (en) | Transformer substation multi-scene inspection analysis method based on federal learning | |
CN110189291A (en) | A kind of general non-reference picture quality appraisement method based on multitask convolutional neural networks | |
CN109218134B (en) | Test case generation system based on neural style migration | |
CN109711449A (en) | A kind of image classification algorithms based on full convolutional network | |
CN110363727A (en) | Image defogging method based on multiple dimensioned dark channel prior cascade deep neural network | |
CN108564166A (en) | Based on the semi-supervised feature learning method of the convolutional neural networks with symmetrical parallel link | |
CN106846261A (en) | Underwater picture processing method based on convolutional neural networks | |
CN107993200A (en) | Picture noise level estimation method based on deep learning | |
CN109241814A (en) | Pedestrian detection method based on YOLO neural network | |
CN110503610A (en) | A kind of image sleet trace minimizing technology based on GAN network | |
CN109191389A (en) | A kind of x-ray image adaptive local Enhancement Method | |
CN109800785A (en) | One kind is based on the relevant data classification method of expression and device certainly | |
CN108537747A (en) | A kind of image repair method based on the convolutional neural networks with symmetrical parallel link | |
CN112686152B (en) | Crop pest identification method with multi-size input and multi-size targets | |
CN108038839A (en) | Twisted-pair feeder lay real-time detection method on a kind of flow production line | |
CN110428011A (en) | A kind of deep learning image fault classification method towards video transmission quality |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191108 |
|
RJ01 | Rejection of invention patent application after publication |